Most reports in the office still look backward. Last month's sales, last quarter's expenses, this week's transaction count. However, the truly determining questions are forward-looking: how much stock should be prepared ahead of Ramadan, can cash flow withstand the burden in the third quarter, when does the demand for a product start to decline.
The gap between these two things is not about tools. Excel is on almost every office computer. What is often missing is the habit, and a bit of skill, to transform piles of old data into accountable projections. Various studies indicate that around 70 percent of the workforce in Indonesia lacks adequate data literacy, while graduates who are truly ready to fill data-based roles are still relatively few each year. This gap is felt precisely at this point.
Why Forecasts Often Miss the Mark Here
Demand patterns in Indonesia are rarely smooth. There are spikes during Ramadan and Lebaran, shopping frenzies on twin dates like 11.11 and 12.12, and consumers who are very sensitive to prices and promotions. Guessing with a straight line is almost certainly wrong. A proper forecasting must take into account seasons, trends, and surprises. At this point, many business players give up and revert to instinct. The problem is, instinct without data is an expensive gamble.
The Tools Are Already in Excel, Just Rarely Opened
Excel holds more forecasting capabilities than are typically used. There is a Forecast Sheet that organizes seasonal projections from just a series of dates and numbers. There are functions like TREND and FORECAST.ETS, moving averages, and the Data Analysis Toolpak for descriptive statistics and correlation. Added to that is Power Query, which can tidy up raw data once and then just needs to be refreshed each week. Unfortunately, most users stop at SUM, VLOOKUP, and a couple of pivots. Its predictive layers are left dormant.
AI as a Data Reading Partner, Not a Magician
This is where AI assistants like Claude start to enter the analyst's workspace. Not to replace human judgment, but to speed up the tedious parts: explaining which forecasting methods are suitable for a pattern, helping to formulate complex formulas, translating a trend into the sentence "so what does it mean" that managers can immediately understand. The decision remains with the person who understands the business context. What changes is the speed from raw data to insight. When most of the workforce is not yet familiar with data analysis, such a bridge is no longer a luxury.
Reading Tomorrow from Yesterday's Numbers
In an economy with thin margins and easily swinging demand, the ability to read tomorrow from yesterday's numbers quietly transforms into a core skill, not just an added value. The spreadsheet open on your screen actually holds more insights than we usually ask of it. The only question is: are we willing to ask further of the data we already have.
References:
- Foreplan – Business Forecasting in Indonesia: Challenges & Solutions → foreplan.id
- Universitas Multimedia Nusantara – Why Learning Data Analytics Is Crucial in 2025 → umn.ac.id
- Majoo – Forecast: Definition, Types, and Examples in Business → majoo.id